I ntroduct ion

This thesis is ahout R villRge society and its relationship with a Hindu

kingship in the 1 ight ()!' the ritual system of an old temple, which includes servic~ tenures granted by the local and the ritual organization based on caste and kinship system in the village. The village; Kshiragram, is famous for being

one of 51 sakta-,-pi thas in , and the village temple (jogadya-mandir) was prosperous under the ratronage of the Haj in the early eighteenth century. 1 The temple history is, however, far older than this as inferred from the mystical legends of the goddess Jogadya recited by the villagers .. · The ritual system of the temple is fundamentally designed for sacrifice. In the medieval period, it is said, there used to be human sacrifice in front of the image of the goddess, which is implied now hy various ritual evidences remained in the ritual courses.

1. The village society and the temple

,_ a) The vi I I age setting The viiage society consists of eighteen castes. But, there are the three major population strenghs: a) the Ugra-Kshatriya, b).the Bagdi, and c) the Brahman. The Ugra-Kshatriya is dominant landholding caste as well as the largest population strengh in the village. They are the main ·patron of the Brahman and the other service castes. They claim the status of Kshatriya caste in the village, and are

"Kshiragram' is the literary name of the village. The popular name is "kshirgram'; however, sometime it is called "kshirgan' colloquially. ln this thesis, I '- adopted the formal name; "kshi rag ram' to refer to the village. -1- called "' colloquially. 2 The Bagdi who constitute the majority among the untouchable castes if! J

Raj who is regarded as the ultimate authority and source of honour to the temple

.ritual organization. The Raj family is the grantor of the temple service lands as well as the temple properties. Moreover, the Maharaja used to be a large who was in possession of many estates including the village bef6re Independence.

There are, therefore, four im~ortant parties included in the temple ritual system;

I) the dominant 1an.dhn l ding caste, 2) the Brahman pri cs ts, :l) the other serv icc castes, and 4)the Maharaja as the supreme authority. They are interdependent with one another through mutual and sometime asymmetrical·relationships such as jajman-

priest (purohit)[l)-2)], patron- service castes [1)-3)], kingship- temple ritual organization [3)·-1),2)], zamindar(landlord)- farmer (raiyat) [1)-3)], and so on.

2 According to the ideal model of the caste system in Bengal, there is neither Kshatriya nor Vaisya [see,e.g.Sarma !980]. Nevertheless, there are a few castes who claim the status of Kshatriya such as the -Rajbangshi in , the Tetuliya-Bagdi in , and the Mahato in Puruliya. The Ugra-Kshatriya is originated from the agricultural caste[Oldham 1894:17]. They mainly inhabit in the district of Bardhaman as the raiyat or the jotdar in rural area. Most of the Ugra-Kshatriya in the villages enjoyed the dominance of the landholdings as well as the status of the uppe'r castes. In this paint, see Risley, H. fl. [ 1981: I ?.-1:3] See a I so, Sanj i b Banchu, Ugrakshatiryn Parici ti. 1986.

I -2- I These relationships could be summarized in brief as kingship, caste, and ritual organization. The particular focuses in the thesis are mainly based on the following three

materials: the first ;s the ritual 11rganization consisted of the ritual p~sts an~ roles of the Jogadya temple, th second is the system of th ritual duties and rights in the temple ritual and rituals in the localities, the third is the ritual processes, in which each caste and lineage is assigned with the various'ritual roles throughout the year. They arc inextricably interwoven with each other around

'·the temple ritual courses. Tt is on these elements that the 1·itual system of the

tempi~ has been based for a long time _and they still continue to sustain it. For the beginning, we shall overview tho village society and the temple based on the historical accounts.

b) Historical background During the late Mugal period, the family of the became a zamindar in the village Baikunthapur, near the town of Bardhaman, under the Mugal empire. Soon they began rapid expansion over the present , and then all over western Bengal. Around the time of the beginning of the British rule, they became the largest zamindar over Bengal in terms of their record of the revenues paid to the government. They exercised great authority over the kingdom in the sense of protection as well as oppression to the people. 3 The Raj family rebuilt the Jogadya temple of today in the village durif!g the reign of the Maharaja Kiritchand(1702-1740). After that, they embarked on building many temples over the kingdom. During the British period, the Maharaja's estate·came under the (l793) and the Raj family became one of the most powerful zamindar under

the British. Government. I L continued to be so until the Estate Acquisition

3 See, in particular, Maclane [1993:,II•:' 69-95]. -3- \ _Act(l953), with two interruptions of' direct control by the court of ward of the'

British government. Under the Bengal Tenancy Act(l985), the rights of owner-cultivators are

prescribed in relation with and it is the first administrative ~ategory which stipulates "raiyat (farmer)" in Bengal. This was a prerequisite in order to transfer the tenurial rights of the raiyats by the intermediator to the direct control under the government. Most of the households of the Ugra-Kshatriya and of the Brahmans in.the village fall undcr'this category. It is also important that the Bengal Tenancy Act provided the operation of the Survey and Settlement of each district over the Beng;1'· :'residency, which is the first attempt to survey each plot of the estates systematically and settle the rents on the record book called "khatiyan" by the British administrators. The rights of tenants and agricultural

labourers, who are mainly scheduled cas~e (untouchable) of today, were practically prescribed after Independence, especially under the Gram Panchayat administration. In this way, the village society consists of the compound relations between the kingship and the village society, the dominant caste and the Brahman or the service castes, the zamindar and the farmers, and so on. This gets reflected particularly in the ritual course of the village temple which occupies the central core of the village society. The field-data collected from the village seems, therefore, suggestive Lo understand village society from the view point of an

indigenous political form~tion. It is the religious aspect of kingship and its impact on village society that were overiooked so far under the influence of the

theory on Hindu . soci eLy hy L. Dumont. In this sense, this monograph intends to reconsider the village social form from the view points of kingship, dominant cas!e, and other service castes in place of the vieW of village society which is based entirely on a religious dichotomy of pure and impure. It will be important at this stage to outline the recent tendency of anthropological studies on South Asia to place the topic for this dissertation clearly in perspective.

-tl- 2. Anthropology and vii lage study. a)anthropology in the 1980s

After the publication of Jfomo lfierarchicus (in 1970 in English) by L.:Dumont, many anthropologists have tried to criticize the theory of his book. They also started paying attention to the theoretical issues regarding the Hinduism. Thus, the village studies of the social structure model became very few in number compared with in the 1950s. Many anthropologists shifted their attention from the village studies to the studies on the religion since 1980s. This fact, however, raises an important question: Docs the decline of the villngc studies necessarily mean the worthlessness of studying in the village any more or does it simply mean .the shortage of the elaborate perspective in the village studies?

According to Fuller & Jonathan[l990], the Lssues regarding Hindu society discussed within the post-Dumontian framework as they refer to it, in the 1980s could be understood through the following three key figures; namely Brahman-priest, ascetic renouncer, and kingship. It could be added some other subjects, such as the rethinking on the concept of pure and impure by Sekine [1989], and the relationship between Hinduism and Hindu society which C.Fuller [1979] himself once discussed. In. the framework of L.Dumont, it is well known 'that the status of the Brahman is fundamentally based on his priestly occupation[l980:47]. Many scholars argue that this opinion is of doubtful acceptabi 1 i ty. For example, J. Parry · [1980] discussed the degenerated status of funeral pries~s or the Mahabrahmans at the cremation ground of Benares, the major sacred centr~ in north India. BecaOse of their permanent role in death ritual and acceptance of dan (gift), they are treated "much like Untouchables" . Fuller [1984] systematically looked into the matter of the Brahman-priests throu~h the analysis of priest organization at the great centre of the llindu pilgrimage, the Minaksi temple of Madurai. He presented enough persuasive ethnographical data which demonstrates the re1ative inferiority of priestly Brahmans to the non-priestly Brahmans.

-5- It was well known that the Kulin-Brahman (kaulinna-pratha-brahman) in Bengal

holding the titles such as. "Mukherjee (Mukho-upaddayya)" , "Banerjee (Bandyo-

upaddayya)" , and "Chatterjee (Choto--upaddayya)" ,, never practiced priesthood, and

their superiority was strengthened by the status as bride-taker to the· other Brahmans[e. g. Risley 1900]. It is ironic that this wife-taking relationship provided a basis for the argument in the marriage alliance theory presented by L. Duman t [1983]. At any rate, as shown by the theoretical study by Quigley [1993:ch.4], it is now widely accepted that the Brahmans are not always priests in Hindu society and often the "ideal" Brahman is not seen as a priest [see, also Burghart 1990; Heestermcn 1985; Van de Veer 1985]. With regard to ascetic renouncers, Dumont thought, as we know, that the

renounced ascetics are only an exc~ptional element in the holistic society of the Hindu, and they correspond to the indiv}duals in western society. This comparison is itself interesting, as this is certainly a comprehensive attempt to contrast Hindu society with western society after Max Weber. However, there is much

evidence which suggests that they are never outside of Hind~ society. Regarding this point, Van der Veer[1989] has observed through the analysis of the

organization of priests ancl n~nounccrs in the Rama temple, the famous pilgrim

centre of Ayodya [n Uttar Pradesh, that religious ascetics never exist independently, and earn their economic assistance from the temple management. Burghart [1983] also argues that there is no simple dichotomy between the Brahman

householders and the renounccrs. Similar studies abound[e. g. Barford '1985; flccstcrman 1985; Parry 19H2]. We have known or similar examples in the tradition of Bengal, particularly, among the Vaisnava and Muslim. Despite their idea of the egalitarian fraternity among castes, the disciples of Caitanya, a famous sect of Vaisnava (baishnab) in Bengal, have reproduced a caste ranking in their sect[Dimock

1966]. It is also wei I knowri that there are caste-like hierarchies in the rural Muslim societies[Sengupta 1973]. The ethnohistorical study such as Vander Veer's· is actually effective in ( verifying the static descriptions based on the so-called "ethnographical present"

-6- such as structural analysis represented by L.Dumont. However, the simple historical

data clearly needs to be distinguished from the ~thnohistorical synthesis; because it sometimes lead to the overemphasis of socio-economic factors and overlook the peculiar religious or ideological factors in the context of Hindu society. This problem is commonly shared not only in South Asian anthropology but also by many \ . contemporary anthropologists. We have tried to avoid oversimplifying both ideological and economical factors since attempts of Levi-Strauss; Geertz, Sahlins,

and so on. To quote Sahlins[l985], "there is no phenomenal ground-let alone a~y

h~uristic advantage- for considering history and structure as exclusive alternative" . In this respect, I shall mention two studies, namely An anthropologist among the Historians. by Cohn [1987] and The Hollow Crown. by Dirks [1987]. The former is based on the anthropological history or the reconstruction of history under the influence of E.Hobsbawm [1983] and it could lead to the field of Sabartan. studies, which is an attempt to collaborate between anthropology and history on Indian societies. The latter is an attempt to synthesis of the structure and history·as ·shown in the study by M.Sahlins [1985], which leads to an emerging field of colonial studies[see, also AppaJurai 1981; Breckenridge & Vander Veer 1993; Inden 1986, 1990; Quigley 1993:ch. l,etc]. In both cases, the approache leads inevitably to the rethinking of history during the colonial period and its impact on the indigenous ~::~~:. political forms of pre-colonial Indi~;~that is the topic of Hindu kingship.

b)Study on kingship During the 1980s, many anthropologists focussed their attention to the study on the Hindu kingship, especially to the studies regarding transformations during the British rule. It can be summarized as follows. At the beginning, the discussions chiefly rnncentrated on the criticism· of the concept of encompassed power to the religious values by Dumont. In opposition to Dumont's opinion about a secularized king, the religious functions of the king were pointed out through examples such as the king' s role as jajman(patronl_vajamana Skt.) of priests, and

-7- his chief position in ~Hcrificial rituals[e. g. Derrett 1976; Fuller 1984:104~6; Heesterman 1995]. On the other hand, Biardeau [1989], taking to the tradition of French Sociology, considers the Brahman-Kshatriya relationship to be the core of the structure of the Hindu society. ;_-According to Biardeau, the king i"s the :~~~... delegate of the gods and the goddesses on the earth and for the protection of:the

kingdom, he plays the role of jajman in the sacrificial rituals performed by,- the Brahman priests. Thus, he thinks the king occupies the primary position in the Hindu society. ... These theoretical queries led to the revaluation of the theory of A.M.Hocart . Quigley [1993:ch.6] argues that it is an alternative to the theory of Dumont and

tries to provide a :•:~]W perspective to the caste society on the basis of a . reinterpretation of Hocart's theory. This attempt further stimulates the reconstruction of another holistic view on Hindu society on behalf of Homo Hierarchicus[Tanabe 1993]. Hocart considers the caste system as a "sacrificial organization" in -·which "castes are merely families to whom various offices in the ritual arc assigned by heredity" [1950:20]. It is worth noting that three aspects of his theory are connected with the subjects discussed in the thesis. First, Hocart suggests that the king or Kshatriya belongs to "the first caste" , because they command other castes to play defini le roles in the rituals in performing sacrifices, ·which .regenerate the cosmological order anil then secure the prosperity of the community. It is worthwhile to mention here that a keen insight into cosmological analogy of _the sacrifice to the world order and prosperity has .been presented in the excellent

book by Hubert & Mauss[l964]. The second important point in his theory .is his emphasis on initiation for caste memhership(ibid. :56]. It seems natural that if the caste system is designed for the sacrificial ritual, then initiation is indispensable, because the descent of caste is not a sufficient condition for each member to participate in tho ritua1. The third point is the idea of the reproduced structures of the kingship in miniature[ibid. :68], which is suggestive enough to

'·- recall the later studies on the local political structures[e.g. Cohn 1962; Stein

-8- 1980], and the images of the polities[e. g. Geertz 1980; Tambiah 1976]. Regarding this fluid political boundaries, Quigley argues: "caste. results when .kingship attempts to assert itse1f' ngainst kinship" [1993:129]. This viewpoint would be useful in explaining a relationship between the caste organization and the kingship. It is also connected with the concept: ~imperial formation" presented by Inden[1990:ch.6], which enables us to see the indigenous polity of India as an incessant generative process.· Another point to the study on kingship concerns the changing circumstances a~6und the kingship under the British rule. As Fuller [1977;1989] acutely pointed out, the self-sufficient village economy, which has so far been understood to refer to the jajmani system, is the result of the colonial administration. Most field workers overlooked the facts that the British government has taken over, especially through the revenue administrations, the position where the Hindu kingships reigned ' over hierarchical polity of the kingdoms in the medieval period. It is, therefore, misleading if anthropologists describe the villages of today as "traditional culture' , because they see the villages which have transformed drastically during about 150 years of the British rule. Inden [1990] considers this matter systematically, and als(J describes the formation of the western views about the Indian society as a dynamic process of interaction between the west and the east. This forms broadly the argument of Oriental ism presented by Asad [1973], Said[l978], and so on. In contrast to the perspectives of Hindu kingship, some scholars pay attention to the problem of the kingship and the dominant caste in village society. Rah.eja' s [1988a] contribution is most remarkable in ·this respect. She suggests that the jajmani relationship is based on the dan (gift) which conveys the inauspiciousness or religious sin from the dominant caste as jajnwn to the ·Brahman priests. It is well known that Oumnnt understands jajmani relations tn be fundamentally encompassed by the concept of pure and impure[.1980]. This enables us to reconsider village society as a central and peripheral system, which consists of the relationship of the dominant caste and the other service castes based on the

-9- idea of auspiciousness and inauspiciousness. This is important as an alternative to the view represented by Dumont in which the village society is hierarchically structured-by an axis hetween the relation of the Brahman and the untouchables based on the idea of pure and impure.

c)Methodological problems Notwithstanding these theoretical advancements anthropologists have failed to accumulate sufficient data concerning the organization of polities and their relationship to the village society frofu the fields of South Asia. There are indeed some reports on the interviews with the ex-Maharaja [Mayer 1981] and the splendid ritual at the former court of' Raj[Ostor 1980]. However, during the long-continued

British rule, the ~~~-.·r_~reignties were practically handed over to the British government and, moreover, the handing over was completed in post-Independence India with the abolition of the princely states in 1971. With the establishment of the modern administrations the indigenous organization of the kingship has been· changed drastically in rural areas while the ritual systems of the villages basically . . controlled by the Brahman priests have remained relatively unchanged. It is under such circumstances that many anthropologists began their field research in the 1950s, This matter has no doubt been appreciated by some anthropologists, but there were very few monographs about it primarily, because of the methodological problems. Since the 1980s, many anthropologists began paying attention to this aspect and they resorted mainly to two methods to tackle these problems. The first is the ethnohistorical approach in which the historical materials preserved in the British offices, such as, state archives and record rooms of the revenue office are· used. The analyses of the historical documents of the colonial rule explain the changing processes of the local political systems at that time and also throw light on the former aspect of the Hindu kingship. It has already been mentioned in the studies by Appadurai [1981], Cohn[1987], Dirks [1987], and such others.

I In the second approach the ritual data preserved in the Hindu sacred places

"""' -10- arc used. These data consist mainly of the ritual systems traditionally followed in the old llindu temples. Formerly, most of the famous temples were supported by the local kings which used to function as religious centres in the respective ' kingdoms. Despite post-Independence changes in their patronage from the :kings, there still remains some old features kept within the ritual systems of these temples. Many anthropologists have tried to reconstruct an outline of the ·relationship between the kings and the priests from these ritual systems. The studios by Fuller[l9H4], Marglin[I9H5], van dcr Voer[l989], and so on, are good examples of this kind. A noticeable trend in the decline of the village studies started in the 1980s as already discussed. Many anthropologists turned from the village studies to the studies on religious centres belonging to the so-called "great traditions" . This tendency seems, however, to be confined to the description of the Hindu society, because these studies have mainly depended on a certain system, which was specifically formed within the rcliRious traditions, and they orten arc quite apart from daily life ·of rural people. For example, Tanaka [1993] shows the case of the Brahman priests of the Nataraja temple in the famous sacred place; Chidambaram in South India where the pr;ast community forms a self-contained society and they are

not dependa~t either economically or ideologically on the patrons. This case is important to provide a counter evidence in opposition to the v1ew that the ideal Brahman has to be free from the subordinative jajmani relations. It is, however, an exceptional case as the Nataraja temple mainly depends on the donations of the

visiting worshippers and is separated from the vill~ge society. This case suggests that within a specified system, the superiority of Brahman can be maintained and at the same time, the data from such particular cases should be distinguished from the observations in village society. It follows from what has heen said that more detailed data on the Hindu kingship during the c,,lonial rule and hefore that is necessary. Jn this respect,; it still seems important to analyze the data in the village enough to overcome "--, these methodological dif"l'iculties, to supplement the historical materials with

-II- them, and to reconstruct from both. In this sense, this synthesis of historical studies and village studies· is never a regression to the field works in the 1950s a:. and a restriction of their objects exclusively into the religion either, It is through a productive combination with feasible varieties of materials that L.Dumont also could construct such an attractive perspective.

3. Methodology ·-· a) Selection of the research vi I I age Selecting a suitable sample village for the purpose of the present study was not an easy task due to the above mentioned methodological problems. It was after a pilot-survey of six villages of that I was able to select the present .village, viz., Kshiragram. The present village was selected considering the

following four conditions; 1) a village which has an old temple managed by a well organized ritual system, 2)a village isolated from abundant modern transportation -facilities thereby turning it into a tourist centre, 3)a village having an adequate

population balance among the dominant landholding caste, the Brahman cas~e, and the unouchables, and 4)a village mainta-ining a close relationship between the village

~- society and the landlord (zamindar). As already mentioned, the village Kshiragram seemed to satisfy all these conditions.

Tt was revealed after I, started visiting the village that the village was

related to the kingship of the Bardhaman Raj even ~efore the Permanent Settlement in many respects, and the temple was under the dtrect control of the Maharaja till Independence. Furthermore, its relationship with the Maharaja continued in the

various ritual processes. 1 t seems, therefore, that it would enable us to

understand how th~ Hindu kingship has been viewed by the villagers and to what extent it has influenced their social life. At the same time, it can provide counterevidences to Dumont's views or village society consisted of the Brahman- untouchable axis.

-12- The pilot survey was done from October, 1992 to March, 1993. During the time, I visited more than one hundred villages in various geographical areas from the frontier villages of .Sundarban to the villages on the foothills of , , and . Through this survey, I found that ritualistic traditions preserved from the pre-colonial lndian tradition have remained in more traditional forms in the areas where there used to be the strong political forms such as Rarh area and Koch rather than in frontier villages. I, then, started to visit the

villages having temples of sakta-pi thas one by one. The reasons were: r First, . the cult of sakta is primarily connected with the rituals related to = sacrifice. Second, in the rituals of Durgapuja, which were for the most part patronized by the Kshatriyas and the kings, the rituals for sacrifice are often concerned with the rituals of kingship. Third, the cult of sakta generally pelongs to the older tradition than the cult of Yaisnava. (baishnab), which is a popular religious sect centering around the town of Nabadwip in Bengal. The cult of Yaisnava is in contrast with the sakta cult and strictly prohibits any killing (ahinsa). Fourth, although there are many places where sacrifices are made in the

puja rituals, the temples of the sakta cult seem to have well organized ritual systems containing of various caste groups in which sacrifices are made through out

the year. Finally, it s(.]emed logical to compare alJ the sakta-pith~s for selecting

the most suitable village for th~ present research purpose. A list of '51 sakta­ pithas is given in Appendix-1.

b) Distribution of sakta-pithas of Bengal According to a .Bengali almanac (pan)ika): P.M.Bagcir Dairektad Pan)ika, there

are eleven sakta-pithas in West Bengal out of fifty one all over the Indian

subcontinent. There ,:1re, however, many local legends about those pithas. For

example, there are twelve sakta-pithas as existing in West B~ngal according to another almanac: !Jenimndhah Siler Fhul PunjikH. Hesides, the West Bengal District

-13- Gazetteers mentioned thirteen pithas. Of thirteen sakta-pithRs, five are reported as existing in the , three in the Bardhaman district, two in Calcutta, one each in. the Medinipur district, Hugli district, and . According to a famous text, the Jfahapitahanirupana, there are .fifteen names of pitha which can be identified to the present pithas in West Bengal. Among them, seven pithas are the names added in the later version as discussed later. But, any classical texts never refer to the actual locations in the mapi and just mention the names of the goddesses and their fallen limbs of the goddess Sati. Hence, it always remains certain possibilities to regard some local shrines as sakta-pitha in connection with the names of pithas on the texts by the local people. For example, the temple of the goddess Melai Candi, the town of Amta, the district, is regarded to be a sakta-pi tha at which, according to the local people, the left shank of th<' goddess Sati fell. However, this pitha is usually thought to be the Jayanti temple in the village Baurbhag, Jayantiya pargana, Srihatt, Bang1 adesh. Appendix-1 gives a list of the fifty one sakta-pithas commonly recognized in ·Bengal. It has been compiled from the almanac: P.1W. Bagcir Dairektari Panjika. This list mentions most of the popular places recognized as sakta-pitha at present. Besides, all pithas of West Bengal mentioned in the various texts and almanacs are shown in Map-1. It is attached a constrative Jist of these sakta-pithas. The

Kalighat temple in Calcutta is ve~y famous, and it is undoubted that the temple of Kalighat has grown with the commercial development of the great city of Calcutta [Roy 1993]. Now, it becomes the great centre for th·e pilgrimages of the Hindu from all over the Indian subcontinent. But, the situations of the other places are very diverse. For examples, the Dakshlneshwar temple in north Calcutta and the Bakreshwar temple in Birbhum are well known as the pilgrim centres in Bengal. The former is very famous in connection with Sri Ramakrishna, the great mystic of Bengal, and the latter is also known for the hot-spring therapy. On the pther hand, the Kalipitha, the jaydurga temple ncar , the Kal i ganj block, the and the

-14- Mangalcandi temple at Kogram, the Mangalkote hlock, the Bardhaman district are visiied only by the neighboring villagers and visitors from 6utside are rare, though these places was prosperous during the medieval period. It is natural that

the temples in places which have developed commercially_ or have other fa~ilities including conveyance attract people from distani places. There are a few pithas which are called by the same name and considered as the same sakta-pi tha. For example, there are two Phullara temples regarded as a sakta-pitha, one is near the- town Lahhpur, at Labhpur Block in Rirbhum District, and the other is at the village of Dakshindhihi, Nirolgram, at Ketugram Block in Bardhaman District. Both these temples have many devout worshippers in the neighboring areas and were under the patronage of the landlords before _Independence. The temple at Kankal i tala near l3olpur in Birbhum District is identified as a pitha with the name "kanct'. lt is interesting that there is another sakta-pitha with the name; Kancipuram near Madras in Tamilnadu, which is famous for as "southern Bcnares" . Furthermore, the Srimadan Gupter Dairektari Panjika identifies the name of "ujani (uJia_vni)" with the famous pilgrimage place ·of Ujjain, in the western part of Madhya Pradesh, where is famous for the great triennial Kumbha Mela festival. 1t is, however, usally identified with the village Kogram, Natunhat at Mangalkotc Block in 13ardhaman District. This place was called UJani in the medieval period, and mentioned in the various Bengali folk literatures called the Mangalkahyii.· ln particu1ar, this place is famous for as the native place of the heroin 13ehula in the Manasamangal.

c) Field work in the vi I I age The village Khsiragrarn is located at the centre of the Kshiragram Gram Panchayat area, the eastern end of the Mangalkote block in the Bardhaman District. The whole village consists three mauzas, namely .JL. No. 127, No. 128, and No. 129 in the Mangalkotc Block. ft takes ahout thirty minutes by hus from the local town

~ of Katwa and two hours by bus from Uurdhaman town. It takes more than six hours

...---·gv -15- 120548 ...... ~ .... H¢3Jf11> 2 ~. Mt\R 1998 by bus to reach there from Bolpur. This includes the time required for changing buses three times and crossing the river Ajoy by bo~t. The total area of the vi I Iage is 1, 268. 0:3 hectares (3, 133. 26 acres) according to 1991 Census. The total irrigated lands amount to 817.86 hectares (2, 020. 9o;acres), unirrigated lands amount to 1161. :~tJ hectares (1, 139.96 acres), and the rest is 200. 17 hectares (494.61 acres). 'There are three primary schools, one high school, and one rural health centre in the village. At the outskirt of the village, there is the Gram Panchayat office. Within the jurisdiction of the panchayat office'there are seven villages. The population of the village, according to 1991 Census, is 4263 (Male 2191, Female 2072) and the number of households is 729. A detailed discussion on population composition will be presented in the next chapter.

I began to go to the village Kshiragram from April, 199:3. But it was after

February, 1994 that I started to stay in the village continuously. Before staying inside the village, T discussed with many persons regarding local administration, parties, colleges and universities concerned with reference to the village and th~

~rea. My first stage of field work in the village continued till May, 1995. After that, l visited in October-November of 1995 and January-February of 1996. d) Descriptfon of the thesis

A preliminary statement about the description style in this monograph see~s important here. The hook, Writing Cui ture by Clifford & Marcus [1986] calls attention t~ the possibility o[ misinterpretations.·caused by the description of ethnographies of status quo based on the so-called "ethnographical present" . Regarding this village, the village temple has undergone drastic transformation five times in the past due to the following historical happenings: 1) the invasions by the Muslims, 2) the rebuilding of the temple by the Maharaja Kirtichand (around

· 1730), 3) the Maratha inv~sions(l742-1751), 4) the Survey and Settlement by the

British government (1927/2~), 5) the abo]ition of the Zamindar(1953). After the criticisms on the approach of structuralism, many scholars began

-16- to pay attention to various social conditions, which could lead to the structural changes. lt is generally agreed now that the description of a certain socicjl process depends on the f"ramcwork of' the interpretation by the researcher. In this monograph, I tried to trace back to the situation in the age of the Maharaja Kirtichand according to the subjects of this study. There is, however, always some doubt about the credibility in the detailed ritual processes, because many parts of them, let alone the social aspect, seem to have transformed already. The data on the present situation were collected by the direct observations and interviews. The resl, which seem to have changed and in some cases even lost after Independence, were s0pplemented by the recollections of the villagers and by other historical materials. 4 In particular, the data on the land tenures of the village have been mainly collected from the land records of 1927(28 preserved in the District Settlement Record Room. As a result, the description of the thesis ( is basically based on the direct observations by the author, but it is also supplemented by other historical data enough to reconstruct the former state of the temple traced back to the 1920s.

e) Composition of the thesis

The composition of the thesis is as follows.' In chapter-I, 1 have discussed two points: i) the religious aspect of the village and its temple, which characterizes the village as a sacred place, and ii) the social background of the village. In chapter-TI; the ritual organization of the temple has been discussed through an examination of·the temple properties an~ the service lands granted by the Maharaja. The ri.tua'l cycle of the temple has been examined in chapter-III with special emphasis on the rotation system of fhe Brahman-sebai t. In chapter-IV, the process of Jogadyapuja, which is the biggest ritual occasion in the village in the

4 In particular, I owed much to the efforts by the historians in the village. It would be fairly impossible to bring this thesis to the present stage without help of following three books: Bardhaman: Itihas o Sanskriti. Sri Jogeshwar Caudhuri, 'Kshiragramer Pracin Oi tijjya.' in Pashcimbangc•r Puja-Parban o Mel a. the late Sri Satya Narayan Mukhopadhyayya, Saktimahapitha Rarer Kshiragram o Debi .fogadya. Sri Sanat Kumar Cakrabartti. All these authors came from the village Kshiragram. -17- lunar month of Baishakh (April-May) has been analyzed, and the organization of the ritual posts has been discussed in this ritual course. In chapter-V, the annual ritual cycle of the vi 11age of past and present are examined, which are analytically divided into three parts, namely the rituals connected wj:th the

temple, the temple and the localities, and the localities. Chapter-VI con~erns the Durgapuja mainly supported by the lineages of the Ugra-Ksha tri yas, and gives a description of a typical rotation and succession system of the Duragapuja by each

lineage. Chapter-VII i~ about Manasapuja, whi~h is mainly supported by the village untouchables(scheduled castes). The Manasa temple in the village is granted to the Bagdi-priests by the Maharaja. But the rituals of the goddess Manasa are closely connected with the ritual organizati.Qn of the .Joagadya temple. Finally, a

discussion on the influence of the kingship on the tempi~ rituals and the village social life and the villagers' way of recognition have been attempted.

' ~.

-18-